back
73 comments
There is a conference in Germany every year called medica. If you are interested in medical devices, I highly recommend going there and checking it out. It's an absolutely MASSIVE grounds with THOUSANDS of vendors showing everything from MRI machines to boroscopes, ambulance equipment...anything related to the medical field.

Something which absolutely surprised me when I was there a few years ago was the amount of IoT health devices available in other markets outside of the US.

There were blood pressure braclets, tons of smart watches that measure health data, connected scales, connected...everything, and TONS of platforms wanted to warehouse this data. My coworker and I were there looking for exactly those sorts of things, and it was great to see so much.

It was actually a bit depressing to see how much stuff is available, just not available in the US. The connected health revolution is definitely coming. Something I was a bit frustrated about this year is how much of this would have been helpful as a covid19 early warning system. Imagine monitoring population-level SPO2 levels, temperature, and activity level.

> and TONS of platforms wanted to warehouse this data. My coworker and I were there looking for exactly those sorts of things, and it was great to see so much.

As I read your comment, I had a growing expectation that you were about to express horror at the mass gathering and warehousing of individuals' health data. I was very surprised when you expressed such excitement at the prospect.

If you acknowledge the threats, you should acknowledge the opportunities as well. Expressing "horror" every time data is mentioned in connection with health data is a very one-sided reflex. Surely, there are challenges, but why not solve them?
> Expressing "horror" every time data is mentioned in connection with health data is a very one-sided reflex.

Good call. It was a mistake for me to post such a once-sided comment.

Completely agree. A lot of suspicion out there; a balance with optimism is important
Your doctor already captures this data.
> It was actually a bit depressing to see how much stuff is available, just not available in the US

One reason for this is strict regulations. U.S. FDA Title 21 CFR Part 11 is very strict as to what a medical device company needs to do, and it's not easy to meet those requirements. Most people do not realize that even in the early startup stage, if you build something that is a medical device, there are rules to follow (mostly about traceability: you have to know what you've built, using what parts, and where those came from).

Shameless plug: I am the solo founder of PartsBox (https://partsbox.com/), which helps a lot with managing traceability when building electronics.

I would be interested in knowing more about the tradeoffs between their system and ours, if you could expand a bit?
Yep. This was my major takeaway. There is so much incredible opportunity there, but the regulators are in the way.
Any idea how this BP device actually works? Is it optical? How is it possible to measure without a cuff?
I guess this is the underlying tech: https://www.leman-micro.com/wp-content/uploads/2020/11/LMD-W...

So the pressure is just measured with a MEMS (microelectromechanical) device, the temperature with a different one, and there's also a pulse oximeter (optical).

Read the article? You pinch it I think. Put it on your forehead for temperature
There have been IoT connected blood pressure monitors, scales, and smart watches with SpO2 monitoring available in the US for years. Check out the products from Withings, Garmin, Apple, etc. What exactly are we missing?
That most of them aren't FDA certified, and can't be used for diagnostic purposes. These things are mostly marketed for "fitness".

(This at least was one of the major hurdles we faced. Maybe the regulations have changed in the last 4 years or so. I'm not in that space anymore. The recommendation to go check out medica is still valid though. It's really cool!)

Sounds amazing. Wondering if you could shed some more light on what's possible these days with IoT health devices? I've been a bit out of the loop on these developments
Unfortunately I haven't been in that space for a while. The parent company to the project we were working on got acquired, and they didn't pursue our project.

We were trying to bring connected health/predictive analytics to a larger, existing healthcare company.

It's just like the condition monitoring and predictive maintenance market, but for humans.
This device monitors pressure needed to cause transient digital arteriolar occlusion using MEMs sensors. It then uses an algorithm that uses this data along with pulse oximetry data to estimate the brachial arterial blood pressure.

In situations where it actually matters ( patients with abnormally low or high BP, patients with low peripheral perfusion, patients with peripheral vasoconstriction, critically ill patients ) this will be measuring something with little or no correlation with the brachial BP.

Theranos made similar promises...

Oh! Skepticism is good. And this type of tech is very good for goal-directed management of fluid resuscitation! Severe peripheral vasoconstriction, not so much, as you mentioned. Really, in my first-hand experience most clinicians use the tech to follow trends in the ICU or surgery.

I wouldn't call the claims Theranos-level, but at the same time I don't see how this is any more useful for the general population than the standard automated blood pressure cuffs. Maybe that will change as things evolve!

This article has a nice review of the history-- https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4312947/.

Exactly my thoughts. It'll work well in well people, and mild outliers, but it's probably [harmful] junk for anybody with any serious conditions.

I guess it might have value if you calibrate it against a good measurement in both arms and then only compare it to historical measurements with the same device. That's to say, it can show you change, which isn't unimportant.

I can't tell for sure from this article, but is this still doing an oscillatory method measurement of MAP?
I have High Blood Pressure. 140/100 when non-medicated during the office day. 110/85-90, medicated and relaxed. Having this since 25 (37 now), my genetics sadly puts a cap on my life expectancy (all other vitals normal, HBP runs in family).

All this to say, I hate measuring my BP and follow this space closely. It's a chore that takes minimum 10 minutes twice a day and I have to largely stop what I'm doing.

I would give so much to a device that could measure my BP as easily as O2, but this device is not it. That picture does not inspire confidence.

All the current "continuous" measurement devices are bulky and inaccurate compared to their larger non-"continuous" counterparts. One day I hope the apple watch can do it, but given the physics of BP doesn't seem realistic.

> It's a chore that takes minimum 10 minutes twice a day

10 minutes sounds like a lot, I have a pretty common OMRON automatic pressure measuring device[0], and it takes about a minute to do the actual measuring, plus, say, another two to take off my sweater and put it back on.

I suppose I am missing something, how come it takes much longer for you?

EDIT: have you tried meditation? As a fellow person with genetic disposition for HBP, it was somewhat helpful for me, when I could do it regularly.

[0] something like this, but older https://omronhealthcare.com/products/5-series-wireless-upper...

>> I suppose I am missing something, how come it takes much longer for you?

Not the OP, but at a guess (one of my relatives has kidney disease and take their blood pressure twice a day every day), it takes about 5' to relax, then another 5' to take two readings and combine the results. My relative takes three readings and keeps the last. From my understanding this is common with electronic devices, where the measured BP changes between readings.

Again from my relative's experience (I helped them with a bit of code to keep track of their BP and plot it etc) the trend is usually monotonic, i.e. either it starts low and goes up a bit, or it starts high and goes down a bit. Sometimes more than "a bit", like from 130/80 to 120/70 (i.e. from hypertensive to normotensive range), etc. Manual sphygmomanometers don't seem to have the same variance in their readings. From my understanding, the digital devices have some kind of model that estimates BP from the readings of an oscillometric sensor and are less accurate than the hand-operated ones. The table-top mercury ones are the most reliable of the bunch. And as to finger-cuff devices, you might as well read tea leaves.

Have you looked at wrist blood pressure meters? Mine takes about 30 seconds.

Wrist usually gives a higher reading than upper arm, but if you use both for a while to get a good idea of what is good on the wrist one you should be able to do most of your daily monitoring by wrist, only needing to do arm when wrist is unexpectedly off.

Expensive, but have you seen this?

https://omronhealthcare.com/products/heartguide-wearable-blo...

> genetics sadly puts a cap on my life expectancy

As it does all of us.

My BP monitor takes 3 reading and reports the average and takes about 2 minutes to do this. What are you doing that takes so long?

I do my best to wait 5 minutes before taking the first reading and 2-3 minutes in-between readings
I got rid of HBP by intermittent fasting, but if your BMI is 20, that is not a method suitable for you.
Just curious, no offense intended: what is your level of fitness and BMI?
None at all. Being ~5'9" & 135 lbs, my back-of-the-napkin BMI is ~20.

Fitness is pretty low these days given COVID, but I used to be quite fit (former Fitness company). That didn't really move my BP much.

My biggest levers for BP are salt intake and work stress. Currently a co-founder so work stress is rather high handled with a bit of mitigation practices like meditation, but definitely not enough.

Overall I'm on borrowed time of sorts and do my best to be thankful for each day... doesn't always come through.

Any papers showing actual clinical trials of this thing?

I ask because the word "clinical", used repeatedly in the article, means... in a clinic. And devices like this have an unfortunate habit of failing to perform in the real world.

One thing which I don't see mentioned is the BLE chipset characteristics.

WuQi does RISC-V BLE chips. I don't know if this product is using that or an ARM.

Someone with better Chinese skills than mine will have to dig at this.

Unfortunately, when you tell me "clinically accurate" but then I see BLE by "super cheap Chinese only" supplier, my skepticism alarm bells start ringing.

This sounds like it might be more accurate than the devices the doctors use now. When my blood pressure is measured it drops so fast the standard device starts beeping and fails to get a fix. Last few times the doctor had to use the hand pump to get a good measurement.

It's probably not super common and obviously psychosomatic, I doubt my blood pressure would drop if I were unconscious, but I imagine if I can make my blood pressure go down that fast, other people might do it less extreme and fool the device.

Anyway I'd be happy to use a less invasive device for sure.

The algorithm for regular NIBP cuffs is interesting. It's actually measuring MAP - mean arterial pressure - and doing a calculation based on that and heart rate to approximate actual systolic and diastolic numbers. This is usually as or more accurate than a manual cuff where those pressures are directly detected (because there is no human error in hearing exactly when blood is flowing again in your artery) but fails badly for the small number of people who fall outside the characteristics of the groups that were used to develop the algorithm that converts MAP and heart rate into systolic/diastolic pressures.

(The gold standard, which all other methods are compared to, is a direct measurement taken by sticking a needle in your artery and putting a pressure gauge on it. For obvious reasons this is avoided as much as possible.)

Betting the only real reason for an ASIC, not a microcontroller, is to make it harder to clone -- presuming the MEMS bit is no different from generic.
I've always enjoyed the sensation of being pressed that the traditional machine generates. A cuffless blood pressure device sounds like it's taking away the best thing the old device had going for it.
Curious if anyone knows what technology is state of art for self service blood taking? (think prick tests where you have to squeeze blood form finger into a vial)
And then a wearable device can correlate your changes in blood pressure with what you're watching on your smartphone, and they can sell that data to advertisers.
IOT medical device security is an increasingly worrying issue, particularly with many platforms wanting to warehouse the data generated.
The accuracy is also often questionable. I have tested cheap smart watches that have heart rate info about as accurate as rolling a dice.
Can someone elaborate on value proposition for this sort of device? What problem(s) does it solve?
How common is high blood pressure? About 1 in 5 adults.[1]

Effects of untreated high blood pressure: loss of mental function, atherosclerosis (so heart attacks and strokes), and vision loss are some of the big ones.[2] It's probably the leading cause of death these days.

So the market for a cheap, reliable and convenient diagnostic device is very large, and wide open at this point.

1. https://health.howstuffworks.com/diseases-conditions/cardiov...

2. https://health.howstuffworks.com/diseases-conditions/cardiov...

I know several folks who worry about managing their blood pressure from hour to hour, this would be as useful for them as no-stick blood sugar monitoring is for diabetics
The main non-invasive way blood pressure is measured by a doctor is by putting a cuff around your upper arm. The cuff has a pump attached to it. The doctor inflates the cuff until the pressure cuts off the blood flow in the main artery of your arm. Then the doctor slowly deflates the cuff and listens for when the blood starts flowing again. The pressure that this happens at is your systolic blood pressure (the larger number). Then they keep deflating and listening until the sound stops. That's your diastolic blood pressure (the smaller number).

A digital blood pressure monitor effectively does the same thing, but without the need for a trained professional. It still has the cuff and the pump. To measure the blood pressure it has to pump and cut off the blood flow to some extent. This obviously makes it inconvenient - it's uncomfortable, it's fairly large, you have to put it around your arm (or wrist or finger), and the pump is usually noisy. The new device seems to be an improvement in these categories.

The main improvement here is size though. Blood pressure monitors usually go around your upper arm and the machine that houses the pump and the electronics is even larger than the cuff. This makes taking frequent measurements difficult. There are smaller devices that go on your wrist, but their accuracy tends to not be great. Even the standard digital blood pressure monitors suffer from inaccuracy.

If the device in the article does what it says and is actually close to as accurate as a regular digital blood pressure monitor, then hopefully we will get more blood pressure data on more people. This should hopefully allow doctors to figure out better treatment plans. However, I find it unlikely that it will match the accuracy of the commonly used digital blood pressure monitors.

In my situation, I have to do home dialysis and need to measure my BP before and after treatment every day. I feel this device would be quicker to use and won't trigger the anxiety that some people feel with a normal BP machine squeezing on their arm. That anxiety can skew your BP numbers considerably.
It's rumored that the next Apple watch will have an infrared glucometer. It would be great if somewhere down the road was a pressure monitor, to complete the basic health check.
If one could get a wearable that monitored blood pressure like a heart rate monitor on watch, that would be quite helpful to see trends and dangerous spikes.
Would love to see opportunities for blood pressure variation over time. Acoustic measures?